ZC3H12D Is a Prognostic Biomarker Associated with Immune Cell Infiltration in Lung Adenocarcinoma

6142

Original Article

ZC3H12D is a prognostic biomarker associated with immune cell infiltration in lung adenocarcinoma

Wangang Gong1,2,3, Wumin Dai1,2,3, Haibin Wei1,2,3, Yongyi Chen1,2,3, Zhiguo Zheng1,2,3

1Institute of Cancer and Basic Medical (ICBM), Chinese Academy of Sciences, Hangzhou, China; 2Zhejiang Cancer Research Institute, Cancer Hospital of University of Chinese Academy of Sciences, Hangzhou, China; 3Zhejiang Cancer Research Institute, Zhejiang Cancer Hospital, Hangzhou, China Contributions: (I) Conception and design: Z Zheng; (II) Administrative support: Z Zheng; (III) Provision of study materials or patients: W Dai, Y Chen; (IV) Collection and assembly of data: W Gong, H Wei; (V) Data analysis and interpretation: W Gong; (VI) Manuscript writing: All authors; (VII) Final approval of manuscript: All authors. Correspondence to: Zhiguo Zheng, PhD. Zhejiang Cancer Research Institute, Zhejiang Cancer Hospital, 1# Banshan Road, Hangzhou 310022, China. Email: [email protected].

Background: ZC3H12 family members have an important role in tumorigenesis and development. However, the relationship between ZC3H12 family members and the prognosis of lung adenocarcinoma (LUAD) and tumor infiltrating lymphocytes is not clear. Methods: The expression of ZC3H12 family members in LUAD was analyzed by UALCAN. UALCAN, Kaplan–Meier Plotter, and GEPIA were used to evaluate the effect of ZC3H12 family members on the prognosis of LUAD. The relationship between prognostic ZC3H12 family members and 14 functional states of LUAD was studied by CancerSEA. The correlation between ZC3H12 and immune cell infiltration was studied by TIMMER. In addition, the correlation between ZC3H12D expression and an immune infiltration gene marker set was analyzed by TISIDB and GEPIA. Finally, the expression of ZC3H12D in LUAD was further verified by the GEO database and immunohistochemical staining. Results: The combined prognostic analysis of UALCAN, Kaplan-Meier Plotter, and GEPIA showed that the up-regulated expression of ZC3H12D mRNA was closely related to an improvement in overall survival rate (OS) in patients with LUAD. There was no significant correlation between ZC3H12D and 14 functional states of LUAD. Further analysis showed that the expression of ZC3H12D was positively correlated with the infiltration of B cells and CD4+T cells in LUAD. The expression of ZC3H12D was also positively correlated with immune markers in LUAD, including B cell-derived TNF and LTA cytokines, CXCL13, and its receptor CXCR5. Immunohistochemical staining showed that the expression of ZC3H12D in LUAD tissue samples was higher than normal lung tissues. Conclusions: These findings suggest that multiple ZC3H12 family members are associated with the prognosis of patients with LUAD tumors. The increased expression of ZC3H12D was correlated with improved prognosis. ZC3H12D was shown to be associated with the level of immune cell infiltration, including B cells and CD4+T cells. Thus, ZC3H12D can be used as a biomarker to judge the prognosis and immune infiltration of LUAD.

Keywords: Bioinformatics analysis; lung adenocarcinoma (LUAD); infiltration of immune cells; ZC3H12D

Submitted Mar 12, 2020. Accepted for publication Aug 05, 2020. doi: 10.21037/tcr-20-1465 View this article at: http://dx.doi.org/10.21037/tcr-20-1465

Introduction carcinoma have a good prognosis (14). Interestingly, members of the ZC3H12 family also provide a local Primary lung cancer is one of the malignancies associated inflammatory environment for tumor development. For with the highest morbidity and mortality worldwide, example, ZC3H12A degrades transcripts of angiogenic especially in China (1). Non-small cell lung cancer factors secreted by cells (VEGF, IL-8, and IL-6) and (NSCLC) is the main pathologic type of lung cancer, downregulation of ZC3H12A induces tumor angiogenesis accounting for approximately 80% of all lung cancers. in renal cell carcinoma endothelial cells (8). Recently, Lung adenocarcinoma (LUAD) is the most common form ZC3H12D has been considered to be a new negative of NSCLC, accounting for more than 50% of cases (2). feedback regulator of Toll-like receptor (TLR) signaling The prognosis of LUAD is poor, with a 5-year survival and macrophage activation. Overexpression of ZC3H12D rate of 4–17% (3). Immunotherapy is a new method for significantly inhibits JNK, ERK, and NF-κB signal the treatment of LUAD, and has achieved unprecedented transduction and macrophage inflammation induced by results in patients with advanced LUAD. However, there TLR2 and TLR4 activation (13). However, the mechanism is a wide range of challenges in clinical application, such as of ZC3H12 family members in LUAD progression and the lack of biomarkers that can effectively predict efficacy, tumor immunology has not been established. prognosis, and treatment-related adverse reactions (4). In this study we used multiple databases to analyze the In addition, a number of studies have shown that tumor- level of expression and prognosis of different ZC3H12 infiltrating lymphocytes, such as tumor-associated macrophages family proteins in patients with LUAD. We also used the (TAM) and tumor-infiltrating neutrophils (TINs), affect the tumor immune research evaluation resource (TIMER) to prognosis and efficacy of immunotherapy (5). Therefore, analyze the correlation between ZC3H12D and LUAD the identification of prognostic biomarkers and new immune cell infiltration to determine the prognostic value immune-related therapeutic targets has enormous potential of ZC3H12D in LUAD and its interaction with tumor in the treatment for LUAD. immune potential. ZC3H12 is a newly discovered zinc finger protein family of the CCCH type, which consists of four ZC3H12A-D members (6). The ZC3H12 protein family, also known Methods as MCPIP, is an endonuclease that degrades mRNA UALCAN data analysis transcripts by recognizing the stem-loop structure at the end of mRNA 3'UTR (7). ZC3H12 family proteins UALCAN (http://ualcan.path.uab.edu) is an easy-to-use are abnormally expressed in a variety of tumors, and interactive portal for in-depth analysis of TCGA gene participate in the development and progression of tumors, expression data. UALCAN can be used to analyze the thus affecting the prognosis. It has been reported that the relative expression of genes in different tumor types and expression of ZC3H12A is down-regulated in breast cancer, grades, and to evaluate the impact of gene expression and neuroblastomas, and renal clear cell carcinomas, and low clinicopathologic features on patient survival. In addition, expression of ZC3H12A is related to tumor grade and poor additional information about the selected gene/target is prognosis for breast cancer (8,9). Studies have shown that provided through connection with HPRD, GeneCards, the expression of ZC3H12C in colon cancer is lower than Pubmed, TargetScan, Human protein atlas, DRUGBANK, normal colon tissues, while overexpression of ZC3H12C Open targets and GTEx. In this study, UALCAN inhibits the migration of colon cancer cells (10). ZC3H12D database was used for the analysis of the expression of was originally considered to be a tumor suppressor, and ZC3H12 family genes in primary LUAD patients and the has also been referred to as p34, MCPIP4, and TFL (11). investigation of the correlation between the ZC3H12 family Recent studies have shown that ZC3H12D reduces the members and the survival of patients. global ubiquitin of cellular proteins, and polyubiquitin plays a key role in promoting the tumor cell proliferation and Kaplan-Meier plotter database analysis survival signal pathway (12,13). It has also been reported that ZC3H12D is highly expressed in human endometrial Kaplan-Meier Plotter analysis was used to assess how carcinoma, but not expressed or weakly expressed in normal 10,461 cancer samples impacted 54,675 genes on survival. tissues, the patients with high expression of endometrial These samples included 5,143 breast, 1,816 ovarian, 2,437

© Translational Cancer Research. All rights reserved. Transl Cancer Res 2020;9(10):6128-6142 | http://dx.doi.org/10.21037/tcr-20-1465 6130 Gong et al. ZC3H12D is a prognostic biomarker in LUAD lung, and 1,065 gastric cancer samples on the HGU133 cistrome.shinyapps.io/timer/). TIMER used the previously Plus 2.0 array. The average follow-up time was 69, 40, 49 published deconvolution statistical method to infer the and 33 months, respectively (15). Kaplan-Meier Plotter abundance of tumor-infiltrating immune cells from gene (http://kmplot.com/analysis/) was used to analyze the expression profiles. The TIMER database included 10,897 relationship between the expression of ZC3H12 family samples from 32 cancer types from The Cancer Genome genes and the survival of LUAD. The hazard ratio (HR) Atlas (TCGA) to estimate the abundance of immune with 95% confidence intervals and log-rank P-value were infiltration. We analyzed the correlation between the also computed. expression of ZC3H12 family genes and the abundance of immune infiltrating cells (including B cells, CD4+T cells, CD8+T cells, neutrophils, macrophages, and dendritic GEPIA was used to verify the correlation between ZC3H12 cells) through a gene module. Gene expression against family genes and LUAD survival tumor purity is displayed on the left-most panel (18). In Gene Expression Profiling Interactive Analysis (GEPIA, addition, the correlation between ZC3H12D expression http://gepia.cancerpku.cn/) is a web-based tool based and gene markers of tumor infiltrating immune cells was on TCGA and GTEx data, including analysis of RNA explored through related modules. The genetic markers sequencing expression data from 9,736 tumors and 8,587 of tumor infiltrating immune cells include CD8+T cells, normal samples from TCGA and GTEx (16). GEPIA T cell (general), B cells, monocytes, tumor associated provides key interactive analysis modules, including tumor/ macrophages, M1 macrophages, M2 macrophages, normal differential expression profile analysis, profile neutrophils, natural killer (NK) cells, dendritic cells (DC), drawing, analysis based on tumor type or pathological stage, T helper cell 1 (Th1) cells, T helper cell 2 (Th2) cells, patient survival analysis, similar gene detection, correlation follicular helper T cell (Tfh) cells. Helper T cell 17 (Th17), analysis and dimensionality reduction analysis, as well as Tregs and T cell depletion). These genetic markers refer rapid customization. We used GEPIA to further verify the to previous studies. The correlation module generates correlation between ZC3H12 expression and clinical stage expression scatter plots in a pair of user-defined genes, as and survival of patients with LUAD. well as Spearman's correlation and statistical significance. Log2 RSEM was used to show the level of gene expression.

Single cell sequencing analysis TISIDB Database analysis CancerSEA (http://biocc.hrbmu.edu.cn/CancerSEA/ http://202.97.205.69/CancerSEA/) is the first dedicated TISIDB (http://cis.hku.hk/TISIDB) is a user-friendly Web database designed to comprehensively explore the different portal that gathers various types of tumor immunology functional states of cancer cells at the single cell level. data resources. It integrates 4,176 records from 2,530 CancerSEA mapped a single cell functional status map publications, including 988 genes related to anti-tumor of cancer, including 14 functional states of 41,900 cancer immunity. In addition, genes related to the resistance cells from 25 cancer types (including stem cell, invasion, or sensitivity of tumor cells to T cell-mediated killing metastasis, proliferation, EMT, angiogenesis, apoptosis, and immunotherapy were identified by analyzing high- cell cycle, differentiation, DNA damage, DNA repair, throughput screening and genomic map data. Third, the hypoxia, inflammation, and aging). CancerSEA facilitates association between any gene and immune function, such determining which functional states are associated with as lymphocytes, immunomodulators and chemokines, genes of interest or a list of genes in different tumors (17). was calculated for 30 TCGA cancer types (19). We used We used CancerSEA to analyze the correlation between TISIDB to analyze the role of ZC3H12D in the tumor- ZC3H12 family genes and 14 functional states in LUAD. immune interaction.

TIMER database analysis Verification of ZC3H12D expression in GEO database

TIMER is a comprehensive database for systematic analysis The gene expression profile was obtained from Gene of immune infiltration of different types of cancer (https:// Expression Omnibus (GEO) database. In this study, five

GEO series (GSE) were used (GSE44077, GSE43458, Statistical analysis GSE19804, GSE18842, and GSE19188). The dataset The statistical analysis survival curve was generated by was selected as follows: the number of (I) samples >100 UALCAN, Kaplan–Meier Plotter, and GEPIA. The results included normal and lung cancer tissues; and the number of of UALCAN, Kaplan–Meier Plotter, and GEPIA are samples in a single group of (II) was >30. The expression of displayed with HR and P or Cox P-values from a log-rank ZC3H12D was analyzed using an unpaired t-test. test along with the values of HR and P or COX P from a logarithmic rank test. The correlation of gene expression Immunohistochemical staining was evaluated by Spearman correlation and statistical significance, and the following absolute thresholds were This study was conducted in accordance with the used to determine the intensity of the correlation: 0.00–0.19, Declaration of Helsinki (as revised in 2013) and it was “very weak”; 0.20–0.39, “weak”; 0.40–0.59, “medium”; conducted with approval from the Ethics Committee of 0.60–0.79, “strong”; and 0.80–1.0, “very strong.” P<0.05 Zhejiang Cancer Hospital (No. IRB-2015-176). Written was considered to be of statistical significance. informed consent was obtained from the participants. The Primary LUAD tissue and corresponding normal lung tissue used in this study were provided by the Biobank of Zhejiang Results Cancer Hospital. The primary antibody was a rabbit anti- Expression of ZC3H12 family members in LUAD human ZC3H12D polyclonal antibody purchased from Proteintech [24991-1-AP]. The immunohistochemical To determine the difference in ZC3H12 family member staining kit was purchased from Beijing Zhongshan Golden expression between LUAD and normal tissues, we used Bridge Biotechnology Co. Ltd. Paraffin-embedded tissue UALCAN to analyze the mRNA expression in ZC3H12 samples were sectioned and placed at 72 for 30 min, then family members between LUAD and normal tissues. The dewaxed and hydrated by xylene and ethanol. Antigens results showed that ZC3H12A was slightly low in LUAD, ℃ were heat retrieved with citrate buffer. After three washes, but there was no statistical significance (P=0.0636). The the slides were placed in 3% hydrogen peroxide solution expression of ZC3H12B and ZC3H12C was significantly for 5–10 min and then washed with running water twice decreased in LUAD, while the expression of ZC3H12D was and PBS for 5 min. The primary antibodies were added significantly higher in LUAD (P<0.001; Figure 1A). At the (1:100 dilution) and incubated overnight at 4 . After same time, we used Expression DIY in GEPIA to analyze washing with water twice and PBS for 5 min, the secondary the expression of ZC3H12 family members in different ℃ antibodies were added and incubated for 20 min, followed pathologic stages of LUAD. The results showed that the by washing, DAB staining, counterstaining, and mounting. expression of ZC3H12D decreased with progression of the disease [Pr (> F) = 0.00274], but there was no significant difference among the other three ZC3H12 family members Evaluation of immunohistochemical staining (Figure 1B). The expression of ZC3H12D detected by immunohistochemistry was evaluated by measuring staining intensity and the Prognostic significance of ZC3H12 family members in proportion of positively-stained cells. The expression LUAD of ZC3H12D in LUAD is localized in the cytoplasm. According to the staining intensity, there were four levels We determined whether the expression of ZC3H12 family including negative (0 point), weak positive (1 point), members was related to the prognosis of LUAD patients. intermediate positive (2 points), and strong positive First, the effect of ZC3H12 family member expression on (3 points). The proportion of positive cells was scored the survival rate was evaluated by UALCAN. The prognosis as follows: <5% = 0 point, 5–25% = 1 point, 26–50% = of LUAD patients with high expression of ZC3H12D 2 points, 51–75% = 3 points, and >75% = 4 points. If the was significantly better than patients with low expression sum of two scores was <4 points, it was considered low of LUAD (P=0.0048), while the expression of the other expression; 0 point indicated not expressed; and ≥4 points three genes was not significantly related to the prognosis of indicated high expression. LUAD (Figure 2A,B,C,D). We used Kaplan–Meier Plotter

A Expression of ZC3H12A in LUAD based on sample types Expression of ZC3H12A in LUAD based on sample types

P=6.3604E-02 P=1.694E-12 60 1.5 50 1.25 40 1.0 30 0.75 20 0.5 10 0.25 0 0 Transcript per million Transcript Transcript per million Transcript −10 −0.25 Normal Primar tumor Normal Primar tumor (n=59) (n=515) (n=59) (n=515)

TCGA samples TCGA samples

Expression of ZC3H12A in LUAD based on sample types Expression of ZC3H12A in LUAD based on sample types

P=1.624E-12 P=1.563E-07 12.5 4

10 3 7.5 2 5 1 2.5 0 0 Transcript per million Transcript Transcript per million Transcript −2.5 −1 Normal Primar tumor Normal Primar tumor (n=59) (n=515) (n=59) (n=515)

TCGA samples TCGA samples

B ZC3H12A ZC3H12B ZC3H12C ZC3H12D 4 8 F value =−0.944 F value =1.66 F value =1.62 F value =4.76 Pr(>F) =0.419 1.5 Pr(>F) =0.176 Pr(>F) =0.183 Pr(>F) =0.00274 3 3 6 1.0 2 2 4 0.5 1 1 2 0.0 0 Stage I Stage I Stage I Stage I Stage II Stage II Stage II Stage II Stage III Stage III Stage III Stage III Stage IV Stage IV Stage IV Stage IV

Figure 1 High ZC3H12D expression in lung adenocarcinoma (LUAD). (A) ZC3H12 family member levels in LUAD according to UALCAN; (B) correlation between ZC3H12D expression and tumor stage in LUAD patients according to GEPIA. to further evaluate the effect of ZC3H12 family member =1.38–2.21, P=3.3e-06). In contrast, the OS of patients with expression on the probability of survival among patients LUAD with high expression of the other three genes was with LUAD. The data analysis based on RNA microarray significantly longer (ZC3H12B: HR =0.67, 95% CI =0.52– showed that high expression of ZC3H12A was significantly 0.86, P=0.0018; ZC3H12C: HR =0.55, 95% CI =0.43–0.71, correlated with poor prognosis (HR =1.75, 95% CI P=2.5e-06; and ZC3H12D: HR =0.52, 95% CI =0.41–0.67,

A P=0.12 B P=0.94 C P=0.14 D P=0.0048 1.00 1.00 1.00 1.00

0.75 Low ZC3H12A 0.75 Low ZC3H12A 0.75 Low ZC3H12A 0.75 Low ZC3H12A High ZC3H12A High ZC3H12A High ZC3H12A High ZC3H12A 0.50 0.50 0.50 0.50

Probability 0.25 Probability 0.25 Probability 0.25 Probability 0.25 n (high) =125 n (high) =128 n (high) =126 n (high) =126 0.00 n (Iow) =377 0.00 n (Iow) =374 0.00 n (Iow) =376 0.00 n (Iow) =376 0 2000 4000 6000 0 2000 4000 6000 0 2000 4000 6000 0 2000 4000 6000 Time in days Time in days Time in days Time in days

E ZC3H12A F ZC3H12B G ZC3H12C H ZC3H12D 1.0 1.0 1.0 1.0 HR =1.75 (1.38−2.21) HR =0.67 (0.52−0.86) HR =0.55 (0.43−0.71) HR =0.52 (0.41−0.67) Logrank P=3.3e-06 Logrank P=0.0018 Logrank P=2.5e-06 Logrank P=3.6e-07 0.8 0.8 0.8 0.8

0.6 0.6 0.6 0.6

0.4 0.4 0.4 0.4

Probability Expression Probability Expression Probability Expression Probability Expression 0.2 0.2 0.2 0.2 High High High High 0.0 Low 0.0 Low 0.0 Low 0.0 Low 0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 Time (months) Time (months) Time (months) Time (months) Number at risk Number at risk Number at risk Number at risk Low 360 203 26 6 1 Low 353 178 47 15 1 Low 336 155 40 16 1 Low 336 162 46 12 1 High 359 144 43 13 0 High 319 166 22 4 0 High 336 189 29 3 0 High 336 182 23 7 0

I J K L 1.0 Low ZC3H12A TPM 1.0 Low ZC3H12B TPM 1.0 Low ZC3H12C TPM 1.0 Low ZC3H12D TPM High ZC3H12A TPM High ZC3H12B TPM High ZC3H12C TPM High ZC3H12D TPM Logrank P=0.091 Logrank P=0.17 Logrank P=0.031 Logrank P=2e-05 0.8 HR (high) =1.3 0.8 HR (high) =0.81 0.8 HR (high) =1.4 0.8 HR (high) =0.52 P (HR) =0.091 P (HR) =0.17 P (HR) =0.031 P (HR) =2.9e-05 0.6 n (high) =239 0.6 n (high) =229 0.6 n (high) =239 0.6 n (high) =237 n (low) =239 n (low) =239 n (low) =237 n (low) =239 0.4 0.4 0.4 0.4 Probability Probability Probability Probability 0.2 0.2 0.2 0.2

0.0 0.0 0.0 0.0 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 Time (months) Time (months) Time (months) Time (months)

Figure 2 OS analysis of ZC3H12 family members in lung adenocarcinoma (LUAD) patients using UALCAN (A,B,C,D), Kaplan-Meier Plotter (E,F,G,H), and GEPIA (I,J,K,L). The P values were calculated using the log-rank test. OS, overall survival.

P=3.6e-07; Figure 2E,F,G,H). Functional state of ZC3H12 family members in LUAD We further verified the prognostic significance of To better understand the role and potential mechanism ZC3H12 family genes in patients with LUAD by using underlying ZC3H12 expression in LUAD, we checked the GEPIA. There was no correlation between the expression functional status of ZC3H12 in LUAD in the CancerSEA of ZC3H12A (HR =1.3, P=0.091) and ZC3H12B (HR database. The data on LUAD in CancerSEA came from =0.81, P=0.17) and OS in LUAD patients. High expression the transcriptome (GSE69405) of individual cancer cells of ZC3H12C (HR =1.4, P=0.031) was significantly from LUAD patients with xenograft tumors (PDX). The correlated with the poor prognosis of LUAD patients. In data includes 3 datasets with a total of 126 single cells. agreement with the survival data of UALCAN and the Tumor cell-enriched PDX cells (LC-PT-45: scRNASeq, Kaplan–Meier Plotter, the high expression of ZC3H12D nasty 34) were analyzed and an additional batch (LC- (HR =0.52, P=2.0e-05) was significantly correlated with Pt-45-Re: scRNA-Seq, nasty 43) was obtained to check the improvement in prognosis of patients with LUAD comparable results. Another Lung cancer PDX case (LC- (Figure 2I,J,K,L). These results confirmed and supported MBT15: scRNA-Seq, nasty 49) was prepared to validate our the prognostic value of ZC3H12D in LUAD. analytic strategy applied in the LC-PT-45 case. The results

© Translational Cancer Research. All rights reserved. Transl Cancer Res 2020;9(10):6128-6142 | http://dx.doi.org/10.21037/tcr-20-1465 6134 Gong et al. ZC3H12D is a prognostic biomarker in LUAD Angiogenesis Apoptosis Cellcycle Differentiation DNAdamage DNArepair EMT Hypoxia Inflammation Invasion Metastasis Proliferation Quiescence Stemness

ZC3H12A

ZC3H12B

ZC3H12C 0.11

0.01

−0.09 ZC3H12D −0.19

−0.30

Figure 3 Correlation between ZC3H12 family members and 14 functional states of lung adenocarcinoma (LUAD) in the CancerSEA database. Red and blue cells indicate positive and negative correlations, respectively. The color intensity is directly proportional to the strength of the correlation. showed that there was no significant correlation between and immune cell infiltration by TISIDB. The expression the ZC3H12 gene family and 14 functional states of LUAD of ZC3H12D was significantly correlated with activated (r<0.4; Figure 3). and immature B cells (Figure S1). These results strongly suggested that ZC3H12D plays a special role in immune cell infiltration of LUAD, especially B cells. Expression of ZC3H12D in LUAD is related to the level of immune infiltration Analysis of the correlation between ZC3H12D and the Tumor-infiltrating lymphocytes affect tumor development immune marker spectrum and patient survival. Therefore, we determined whether the expression of ZC3H12 family members was related To study the correlation between ZC3H12D and immune to the level of immune infiltration in patient with LUAD. cell infiltration, we used TIMMER and GEPIA to We evaluated the correlation between the expression analyze the correlation between ZC3H12D and a variety of ZC3H12 family members and the level of immune of immune cell markers in LUAD with ZC3H12A as infiltration. Among the four genes in the ZC3H12 family, a control. We analyzed the immune marker genes of with the exception of ZC3H12C, the expression of the ZC3H12D and different immune cells, including CD8+ other three genes was significantly correlated with tumor T cells, T cells (general), B cells, monocytes, TAMs, M1 purity. Only the expression of ZC3H12D was positively and M2 macrophages, neutrophils, NK cells and DCs with correlated with the infiltration of B and CD4+ T cells (r>0.4, ZC3H12A as a control. At the same time, we also analyzed P<0.05; Figure 4). At the same time, we further verified the different functional T cells, such as Th1 cells, Th2 cells, correlation between the expression of ZC3H12D in LUAD Tfh cells, Th17 cells, Tregs, and depleted T cells (Table 1).

Expression level (log2 RSEM)

ZC3H12A

ZC3H12B

ZC3H12C

ZC3H12D

Intfiltrtion level

Figure 4 Correlation of ZC3H12 family member expression with immune infiltration level in lung adenocarcinoma (LUAD). The expression of ZC3H12D was negatively correlated with tumor purity, and positively correlated with the infiltration of B and CD4+T cells in LUAD (r>0.4, P<0.001).

Interestingly, most of the B cell markers were associated ZC3H12D expression verification with ZC3H12D expression. Specifically, the expression The datasets that met the criteria were filtered through of ZC3H12D was significantly correlated with the B cell the GEO database. There were five datasets that met the markers, CD79 and CD19 (P<0.001). We used GEPIA criteria, as follows: GSE18842; GSE19188; GSE19804; to verify the correlation between ZC3H12D and the GSE43458; and GSE44077. There were 332 cases of aforementioned markers (Table 2). At the same time, we NSCLC and 266 normal controls. The results of a non- also analyzed the correlation between the expression of paired t-test showed that the expression of ZC3H12D ZC3H12D and immunoinhibitors, immunostimulators, was significantly higher in 286 cases of NSCLC in the and MHC molecules. The expression of ZC3H12D was GSE19188, GSE19804, GSE43458, and GSE44077 datasets positively correlated with immunostimulators, including than 221 cases in the control group, while ZC3H12D was CD27, CD28, CD40LG, and LTA, and MHC molecules, highly expressed in tumor groups in the GSE18842 dataset, including HLA-DOA, HLA-DOB, HLA-DQA1, and but the difference was not statistically significant Figure( S2). HLA-DPB1 (Figure 5). In addition, the correlation We then analyzed 10 pairs of LUAD tissues and the between ZC3H12D and immune-related chemokines and corresponding normal lung tissues to verify ZC3H12D receptors was evaluated. The expression of ZC3H12D was expression. IHC staining showed that ZC3H12D was significantly correlated with the chemokine, CXCL13, and abundantly and uniformly expressed in tumor samples but its receptor (CXCR5) derived from B cells (Figure 6). was significantly down-regulated in all adjacent non-tumor

Table 1 The correlation of ZC3H12D and ZC3H12A with related genes and immune cell markers in TIMER was analyzed ZC3H12D ZC3H12A

Description Gene markers None Purity None Purity

Cor P Cor P Cor P Cor P

CD8+ T cell CD8A 0.476 *** 0.376 *** 0.048 0.281 −0.042 0.348

CD8B 0.427 *** 0.348 *** 0.038 0.391 −0.024 0.591

CD3D 0.572 *** 0.473 *** 0.122 ** 0.025 0.573

CD3E 0.631 *** 0.546 *** 0.106 * −0.005 0.907

CD2 0.622 *** 0.533 *** 0.114 ** 0.011 0.806

B cell CD19 0.618 *** 0.546 *** 0.099 * 0.003 0.955

CD79A 0.545 *** 0.462 *** 0.072 0.101 −0.019 0.680

Monocyte CD86 0.453 *** 0.332 *** 0.088 0.0457 0.005 0.908

CSF1R 0.46 *** 0.352 *** 0.131 ** 0.067 0.138

TAM CCL2 0.156 *** 0.019 0.669 0.199 *** 0.132 **

CD68 0.3 *** 0.186 *** 0.111 * 0.042 0.351

IL10 0.381 *** 0.265 *** 0.106 * 0.033 0.463

Neutrophils CEACAM8 0.135 ** 0.123 ** −0.06 0.171 −0.064 0.154

ITGAM 0.42 *** 0.32 *** 0.154 *** 0.093 0.0394

CCR7 0.655 *** 0.576 *** 0.144 ** 0.05 0.268

KIR2DL1 0.268 *** 0.224 *** 0.046 0.3 0.013 0.780

KIR2DL3 0.234 *** 0.159 *** 0.074 0.0949 0.036 0.429

KIR2DL4 0.24 *** 0.179 *** 0.115 0.0917 0.064 0.0833

KIR3DL1 0.238 *** 0.181 *** 0.101 0.0213 0.075 0.0947

KIR3DL2 0.324 *** 0.255 *** 0.026 0.553 0.017 0.661

KIR3DL3 0.165 *** 0.15 *** 0.029 0.506 0.019 0.675

KIR2DS4 0.243 *** 0.184 *** 0.031 0.476 −0.001 0.985

Dendritic cell HLA-DPB1 0.457 *** 0.356 *** 0.098 0.0269 0.03 0.512

HLA-DQB1 0.3461 *** 0.241 *** 0.139 ** 0.084 0.0619

HLA-DRA 0.405 *** 0.293 *** 0.082 0.0639 0.013 0.773

HLA-DPA1 0.418 *** 0.318 *** 0.057 0.195 −0.009 0.841

CD1C 0.329 *** 0.239 *** 0.076 0.0848 0.024 0.593

NRP1 0.096 * 0.047 0.293 −0.024 0.586 −0.047 0.300

CD11c 0.54 *** 0.455 *** 0.28 *** 0.224 ***

Th1 TBX21 0.544 *** 0.457 *** 0.139 ** 0.046 0.306

STAT4 0.428 *** 0.316 *** 0.204 *** 0.119 **

STAT1 0.244 0.121 0.14 ** 0.057 0.195 −0.012 0.784

IFNG 0.35 * 0.263 *** 0.123 ** 0.066 0.143

TNF-α 0.446 *** 0.335 *** 0.086 * 0.297 ***

Table 1 (continued)

Table 1 (continued) ZC3H12D ZC3H12A

Description Gene markers None Purity None Purity

Cor P Cor P Cor P Cor P

Th2 GATA3 0.333 *** 0.202 *** 0.086 * −0.006 0.889

STAT6 0.195 *** 0.217 *** 0.179 *** 0.176 ***

STAT5A 0.5 *** 0.396 *** 0.183 *** 0.117 ***

IL13 0.191 *** 0.123 ** 0.183 *** 0.117 **

Tfh BCL6 0.121 ** 0.129 ** 0.183 *** 0.14 **

IL21 0.35 *** 0.303 *** 0.028 0.523 −0.008 0.857

Th17 STAT3 0.141 ** 0.162 *** −0.056 0.206 −0.05 0.270

IL17A 0.235 *** 0.17 *** 0.17 *** 0.14 **

Treg FOXP3 0.573 *** 0.481 *** 0.146 *** 0.047 0.300

CCR8 0.589 *** 0.501 *** 0.063 0.153 −0.025 0.577

STAT5B 0.416 *** 0.409 0.593 −0.044 0.316 −0.06 0.183

TGFB1 0.245 *** 0.142 *** 0.159 *** 0.099 *

Mast TLR2 0.254 *** 0.169 *** 0.08 0.0713 0.0464 0.304

TLR4 0.441 *** 0.322 *** 0.009 0.843 −0.08 0.0776

TPSB2 0.191 *** 0.078 0.0839 0.131 ** −0.225 ***

CD33 0.423 *** 0.312 *** 0.119 ** 0.054 0.231

CD44 0.214 *** 0.085 *** 0.123 ** 0.07 0.119

T cell exhaustion PDCD1 0.536 *** 0.454 ** 0.212 *** 0.144 **

CTLA4 0.632 *** 0.558 *** 0.229 *** 0.154 ***

LAG3 0.426 *** 0.348 *** 0.18 *** 0.128 **

HAVCR2 0.408 *** 0.281 *** 0.11 * 0.026 0.560

GZMB 0.344 *** 0.243 *** 0.157 0.815 0.097 * *, P<0.05; **, P<0.01; ***, P<0.001.

lung tissues (Figure 7). The clinicopathological information (20-22); however, the biological analysis of the ZC3H12 of patients with LUAD used in immunohistochemical is family in LUAD has not been reported. This study analyzed shown in Table S1. In general, like the TCGA database the expression and prognosis of ZC3H12 family members based on RNA high-throughput sequencing, ZC3H12D in LUAD. ZC3H12D was significantly up-regulated in was also highly expressed in chip-based datasets and was early LUAD tissues, and the patients with high expression highly expressed in LUAD tissues, which can be used as a of LUAD had a good prognosis. Independent databases marker of LUAD prognosis and immune cell infiltration. (UALCAN, Kaplan-Meier Plotter and GEPIA analysis) were used to analyze the prognosis of ZC3H12 family members in patients with LUAD. Due to the difference in Discussion number of LUAD patients included in the survival statistics The relationship between ZC3H12 family dysfunction and the threshold of high and low classification, the results and a variety of cancers has been partially reported of the databases were different, but together the three

Table 2 Analysis of the correlation between ZC3H12D and ZC3H12A and CD8+ T cell and B cell related genes and markers in GEPIA ZC3H12D ZC3H12A Description Gene marker R P R P

CD8+ T cell CD8A 0.44 *** 0.032 0.48

CD8B 0.38 *** −0.0039 0.93

CD3D 0.55 *** 0.086 0.059

CD3E 0.62 *** 0.1 *

CD2 0.62 *** 0.094 *

B cell CD19 0.58 *** 0.11 *

CD79A 0.48 *** 0.027 0.55 *, P<0.05; ***, P<0.001.

C10orf54 A CD27 B B2M CD276 CD28 HLA-A CD40 CD40LG CD48 HLA-B CD70 CD80 HLA-C CD86 CXCL12 HLA-DMA CXCR4 ENTPD1 HLA-DMB HHLA2 ICOS HLA-DOA ICOSLG IL2RA 1 HLA-DOB 1 IL6 IL6R HLA-DPA1 KLRC1 KLRK1 HLA-DPB1 0 LTA MICB HLA-DQA1 NT5E PVR HLA-DQA2 RAET1E TMEM173 TMGD2 HLA-DQB1 TNFRSF13B HLA-DRA TNFRSF13C −1 −1 TNFRSF14 TNFRSF17 HLA-DRB1 TNFRSF18 TNFRSF25 HLA-E TNFRSF4 TNFRSF8 HLA-F TNFRSF9 TNFSF13 HLA-G TNFSF13B TNFSF14 TAP1 TNFSF15 TNFSF18 TAP2 TNFSF4 TNFSF9 TAPBP ULBP1 OV OV LGG LGG KIRP LIHC KIRP LIHC GBM GBM KIRC KIRC KICH KICH STAD STAD PAAD PAAD BLCA BLCA TGCT TGCT ESCA ESCA READ READ LUSC LUSC THCA THCA PRAD PRAD SARC SARC CESC CESC LLIAD LLIAD BRCA BRCA CHOL CHOL PCPG UCEC PCPG UCEC HNSC HNSC COAD COAD MESO MESO SKCM SKCM

C LUAD (517 samples) D LUAD (517 samples) E LUAD (517 samples) F LUAD (517 samples) 5.0 2.5 2 5.0 2.5 0.0 0 2.5 −2.5 −2 0.0 CD28_exp

0.0 LTA_exp CD40LG_exp

CD27_exp −4 −5.0 −2.5 −2.5 −6 −6 −4 −2 0 2 −6 −4 −2 0 2 −6 −4 −2 0 2 −6 −4 −2 0 2 ZC3H12D_exp ZC3H12D_exp ZC3H12D_exp ZC3H12D_exp rho =0.616, P<2.2e-16 rho =0.604, P<2.2e-16 rho =0.621, P<2.2e-16 rho =0.601, P<2.2e-16

G LUAD (517 samples) H LUAD (517 samples) I LUAD (517 samples) J LUAD (517 samples) 11 10.0 7.5 5.0 7.5 9 2.5 5.0 5.0 7 0.0 HLA-DPB1_exp

−2.5 HLA-DQA1_exp HLA-DOA_exp 2.5 5 HLA-DOB_exp −2.5 −6 −4 −2 0 2 −6 −4 −2 0 2 −6 −4 −2 0 2 −6 −4 −2 0 2 ZC3H12D_exp ZC3H12D_exp ZC3H12D_exp ZC3H12D_exp rho =0.456, P<2.2e-16 rho =0.469, P<2.2e-16 rho =0.434, P<2.2e-16 rho =0.42, P<2.2e-16

Figure 5 Correlation of ZC3H12D expression with immunostimulators and MHC molecules in LUAD by TISIDB. (A,B) Correlation between the expression of ZC3H12D and immunostimulators and MHC molecules. (C,D,E,F) The expression of ZC3H12D was positively correlated with immunostimulators, including CD27, CD28, CD40LG, and LTA. (G,H,I,J) The expression of ZC3H12D was positively correlated with MHC molecules, such as HLA-DOA, HLA-DOB, HLA-DQA1, and HLA-DPB1 (r>0.4, P<0.001).

A B

CCL1 CCL2 CCR1 CCL3 CCL4 CCR2 CCL5 CCL7 CCR3 CCL8 CCL11 CCR4 CCL13 CCL14 CCL15 CCR5 CCL16 CCL17 CCR6 CCL18 CCL19 1 CCR7 1 CCL20 CCL21 CCR8 CCL22 CCL23 0 CCL24 CCR9 CCL25 CCL26 CCR10 CCL27 CCL28 CXCR1 CX3CL1 CXCL1 CXCL2 CXCR2 −1 −1 CXCL3 CXCL5 CXCR3 CXCL6 CXCL8 CXCR4 CXCL9 CXCL10 CXCR5 CXCL11 CXCL12 CXCL13 CXCR6 CXCL14 CXCL16 XCR1 CXCL17 XCL1 CX3CR1 XCL2 OV OV LGG LGG KIRP LIHC GBM KIRC KIRP LIHC KICH GBM KIRC KICH STAD STAD PAAD BLCA TGCT ESCA LUAD READ LUSC PAAD THCA PRAD BLCA SARC CESC TGCT ESCA READ LUSC BRCA THCA CHOL PRAD PCPG UCEC SARC CESC LLIAD HNSC BRCA COAD CHOL PCPG UCEC MESO HNSC SKCM COAD MESO SKCM

C LUAD (517 samples) D LUAD (517 samples)

10 6

3 5 CXCL13_exp

CXCR5_exp 0

−3

−6 −4 −2 0 2 −6 −4 −2 0 2 ZC3H12D_exp ZC3H12D_exp

Figure 6 Correlation of ZC3H12D expression with immunostimulators and MHC molecules in LUAD by TISIDB. (A,B) Correlation between the expression of ZC3H12D and chemokines and receptors. (C,D) The expression of ZC3H12D was significantly correlated with the chemokine, CXCL13, and its receptor CXCR5 derived from B cells (r>0.4, P<0.001). databases showed that there was a positive correlation and expression of ZC3H12D was significantly correlated with consistency between ZC3H12D and LUAD prognosis. the level of LUAD B cell infiltration and CD4+T cell In addition, we also found that there was no significant infiltration. The correlation between ZC3H12D expression correlation between ZC3H12D and 14 functional states of and immunocyte marker genes suggests that ZC3H12D the tumor. Our results provide important insight into the plays an important role in tumor immune regulation lack of ZC3H12D expression as a poor prognostic marker of LUAD. Specifically, ZC3H12D was significantly in LUAD. correlated with the B cell markers, CD79 and CD19. The Another important finding in this study was that the expression of ZC3H12D was positively correlated with

Case 1 Case 2 Case 3 Case 4 Case 5

Tumor

Normal

Case 6 Case 7 Case 8 Case 9 Case 10

Tumor

Normal

Figure 7 ZC3H12D immunohistochemistry in lung adenocarcinoma (LUAD) tissues and adjacent normal lung tissues. immunostimulators, including CD27, CD28, CD40LG, (LT) through TLR4 signals. Moreover, TLR4 signals are and LTA, and MHC molecules, including HLA-DOA, positive feedback loops to support TLS formation and high HLA-DOB, HLA-DQA1, and HLA-DPB1. The expression density (26). Some studies have shown that CXC13-coupled of ZC3H12D was positively correlated with cytokines, CpG-ODN stimulation of CXCR5-expressing B cells including TNF and LTA, CXCL13, and its receptor trigger the killing effect of CD8+T cells, thus eliminating CXCR5. lung metastasis in 4T1.2 tumor-bearing mice (27). B cells Recent studies have explained the possible mechanism of with a tumor-promoting effect are defined as regulatory B ZC3H12D expression associated with immune infiltration cells (Bregs). Bregs secreting cytokines, such as IL-10 and and good prognosis. Tumor-infiltrating B cells (TIBs) are TGF-β, inhibit Th1/CD8+T cell function and promote important regulatory cells in the progression of lung cancer tumor proliferation (28). Some Bregs have a CD19+CD5+ and has a dual role in the tumor microenvironment (23). phenotype driven by STAT3 activation and overexpression It has been reported that the number of TIBs is positively of CD5 induced by IL-6, which is thought to promote correlated with the prognosis of NSCLC, especially in tumor progression by inducing tumor angiogenesis and LUAD (24). TIBs promote T cell responses by secreting immunity (29). Some studies have shown that ZC3H12D cytokines and chemokines, maintains the neutralization is involved in the degradation of pro-inflammatory mRNA, function of the tertiary lymphoid structure (TLS) of such as mRNA of IL-6, IL-10, and IER3 (30). ZC3H12D lung cancer, and has an anti-tumor immune effect (25). leads to a decrease in IL-10 and IL-6, inhibits Bregs, and For example, B cells produce CXCL13 and lymphotoxin exhibits an anti-tumor effect.

In conclusion, the increase of ZC3H12D expression is CA Cancer J Clin 2011;61:69-90. related to the good prognosis of LUAD and the increase 2. Testa U, Castelli G, Pelosi E. Lung Cancers: Molecular in B cells. Therefore, ZC3H12D may play an important Characterization, Clonal Heterogeneity and Evolution, role in LUAD immune cell infiltration as a biomarker of and Cancer Stem Cells. Cancers (Basel) 2018;10:248. infiltration and prognosis. 3. Harada H, Murayama S. Proton beam therapy in non- small cell lung cancer: state of the art. Lung Cancer (Auckl) 2017;8:141-5. Acknowledgments 4. Camidge DR, Doebele RC, Kerr KM. Comparing and Funding: Our study was supported financially by the Natural contrasting predictive biomarkers for immunotherapy Science Foundation of Zhejiang Province (LY19H160003) and targeted therapy of NSCLC. Nat Rev Clin Oncol and the Medical Health Science and Technology Project 2019;16:341-55. of Zhejiang Provincial (2018KY294). The specimens were 5. Kalbasi A, Ribas A. Tumour-intrinsic resistance to immune supported by the Zhejiang Cancer Hospital Biospecimen checkpoint blockade. Nat Rev Immunol 2020;20:25-39. Repository and National Human Genetic Resource Sharing 6. Liang J, Wang J, Azfer A, et al. A novel CCCH-zinc finger Service Platform (2005DKA21300). protein family regulates proinflammatory activation of macrophages. J Biol Chem 2008;283:6337-46. Footnote 7. Fu M, Blackshear PJ. RNA-binding proteins in immune regulation: a focus on CCCH zinc finger proteins. Nat Data Sharing Statement: Available at http://dx.doi. Rev Immunol 2017;17:130-43. org/10.21037/tcr-20-1465 8. Marona P, Gorka J, Mazurek Z, et al. MCPIP1 Downregulation in Clear Cell Renal Cell Carcinoma Conflicts of Interest: All authors have completed the ICMJE Promotes Vascularization and Metastatic Progression. uniform disclosure form (available at http://dx.doi. Cancer Res 2017;77:4905-20. org/10.21037/tcr-20-1465). The authors have no conflicts 9. Lu W, Ning H, Gu L, et al. MCPIP1 Selectively of interest to declare. Destabilizes Transcripts Associated with an Antiapoptotic Gene Expression Program in Breast Cancer Cells That Ethical Statement: The authors are accountable for all Can Elicit Complete Tumor Regression. Cancer Res aspects of the work in ensuring that questions related 2016;76:1429-40. to the accuracy or integrity of any part of the work are 10. Suk FM, Chang CC, Lin RJ, et al. MCPIP3 as a Potential appropriately investigated and resolved. This study was Metastasis Suppressor Gene in Human Colorectal Cancer. conducted in accordance with the Declaration of Helsinki Int J Mol Sci 2018;19:1350. and it was conducted with approval from the Ethics Committee of Zhejiang Cancer Hospital (No. IRB-2015- 11. Wang M, Vikis HG, Wang Y, et al. Identification of a 176). Written informed consent was obtained from the novel tumor suppressor gene p34 on human chromosome participants. 6q25.1. Cancer Res 2007;67:93-9. 12. Chan CH, Jo U, Kohrman A, et al. Posttranslational Open Access Statement: This is an Open Access article regulation of Akt in human cancer. Cell Biosci 2014;4:59. distributed in accordance with the Creative Commons 13. Huang S, Qi D, Liang J, et al. The putative tumor Attribution-NonCommercial-NoDerivs 4.0 International suppressor Zc3h12d modulates toll-like receptor signaling License (CC BY-NC-ND 4.0), which permits the non- in macrophages. Cell Signal 2012;24:569-76. commercial replication and distribution of the article with 14. Wakahashi S, Kawakami F, Wakahashi K, et al. the strict proviso that no changes or edits are made and the Transformed Follicular Lymphoma (TFL) Predicts original work is properly cited (including links to both the Outcome in Advanced Endometrial Cancer. Cancer formal publication through the relevant DOI and the license). Epidemiol Biomarkers Prev 2018;27:963-9. See: https://creativecommons.org/licenses/by-nc-nd/4.0/. 15. Gyorffy B, Lanczky A, Szallasi Z. Implementing an online tool for genome-wide validation of survival-associated biomarkers in ovarian-cancer using microarray data from References 1287 patients. Endocr Relat Cancer 2012;19:197-208. 1. Jemal A, Bray F, Center MM, et al. Global cancer statistics. 16. Tang Z, Li C, Kang B, et al. GEPIA: a web server

for cancer and normal gene expression profiling and histological type and smoking habit in completely resected interactive analyses. Nucleic Acids Res 2017;45:W98-102. non-small-cell lung cancer. Ann Oncol 2016;27:2117-23. 17. Yuan H, Yan M, Zhang G, et al. CancerSEA: a cancer 25. Wang SS, Liu W, Ly D, et al. Tumor-infiltrating B cells: single-cell state atlas. Nucleic Acids Res 2019;47:D900-8. their role and application in anti-tumor immunity in lung 18. Li T, Fan J, Wang B, et al. TIMER: A Web Server for cancer. Cell Mol Immunol 2019;16:6-18. Comprehensive Analysis of Tumor-Infiltrating Immune 26. Litsiou E, Semitekolou M, Galani IE, et al. CXCL13 Cells. Cancer Res 2017;77:e108-10. production in B cells via Toll-like receptor/lymphotoxin 19. Ru B, Wong CN, Tong Y, et al. TISIDB: an integrated receptor signaling is involved in lymphoid neogenesis in repository portal for tumor-immune system interactions. chronic obstructive pulmonary disease. Am J Respir Crit Bioinformatics 2019;35:4200-2. Care Med 2013;187:1194-202. 20. Ligeza J, Marona P, Gach N, et al. MCPIP1 contributes to 27. Bodogai M, Lee Chang C, Wejksza K, et al. Anti-CD20 clear cell renal cell carcinomas development. Angiogenesis antibody promotes cancer escape via enrichment of tumor- 2017;20:325-40. evoked regulatory B cells expressing low levels of CD20 21. Labedz-Maslowska A, Lipert B, Berdecka D, et al. and CD137L. Cancer Res 2013;73:2127-38. Monocyte Chemoattractant Protein-Induced Protein 1 28. Taghavi N, Mohsenifar Z, Baghban AA, et al. CD20+ (MCPIP1) Enhances Angiogenic and Cardiomyogenic Tumor Infiltrating B Lymphocyte in Oral Squamous Potential of Murine Bone Marrow-Derived Mesenchymal Cell Carcinoma: Correlation with Clinicopathologic Stem Cells. PloS One 2015;10:e0133746. Characteristics and Heat Shock Protein 70 Expression. 22. Skalniak A, Boratyn E, Tyrkalska SD, et al. Expression of Patholog Res Int 2018;2018:4810751. the monocyte chemotactic protein-1-induced protein 1 29. Zhang C, Xin H, Zhang W, et al. CD5 Binds to decreases human neuroblastoma cell survival. Oncol Rep Interleukin-6 and Induces a Feed-Forward Loop with 2014;31:2385-92. the Transcription Factor STAT3 in B Cells to Promote 23. Patel AJ, Richter A, Drayson MT, et al. The role of B Cancer. Immunity 2016;44:913-23. lymphocytes in the immuno-biology of non-small-cell 30. Wawro M, Kochan J, Krzanik S, et al. Intact NYN/ lung cancer. Cancer Immunol Immunother 2020;69:325-42. PIN-Like Domain is Crucial for the Degradation of 24. Kinoshita T, Muramatsu R, Fujita T, et al. Prognostic value Inflammation-Related Transcripts by ZC3H12D. J Cell of tumor-infiltrating lymphocytes differs depending on Biochem 2017;118:487-98.

Cite this article as: Gong W, Dai W, Wei H, Chen Y, Zheng Z. ZC3H12D is a prognostic biomarker associated with immune cell infiltration in lung adenocarcinoma. Transl Cancer Res 2020;9(10):6128-6142. doi: 10.21037/tcr-20-1465

Figure S1 Correlation between the expression of ZC3H12D in lung adenocarcinoma (LUAD) and immune cell infiltration by TISIDB. Figure S2 The expression of ZC3H12D was significantly higher in 286 cases of NSCLC than 221 controls in the 4 datasets (GSE19188, GSE19804, GSE43458, and GSE44077), while the expression of ZC3H12D was higher in the GSE18842 dataset, but not statistically significant. Table S1 Clinicopathological information of patients with lung adenocarcinoma (LUAD) used in immunohistochemical Tumor Lymph node Smoking Vascular tumor Visceral pleura Cases Gender Age, years Differentiation diameter, cm metastasis index thrombus involvement

1 Female 62 6×5×4.5 Yes 0 Moderately-low No No

2 Female 50 4.5×3.7×2.6 Yes 0 – Yes Yes

3 Female 53 2.3×2×2 No 0 – No Yes

4 Female 64 2.7×2×1.5 Yes 0 – Yes Yes

5 Male 52 7×6×4 Yes >20 Moderately-low Yes Yes

6 Female 79 2×1.8×1.5 Yes >20 – No Yes

7 Male 58 3.3×2.5×2 No >20 Moderately-low Yes No

8 Male 66 4×3.8×2.4 Yes >20 – No Yes

9 Male 53 2×1.5×1 Yes >20 Moderately-low Yes Yes

10 Female 59 2×1.5×1 No 0 Moderately No Yes